Economy and Society Volume 34 Number 3 August 2005: 389
/403

The social relations of

money as universal
equivalent: a response to
Ingham

Costas Lapavitsas

Abstract

In a recent debate in Economy and Society, Ingham criticized Marxist theory of

money on the grounds that it relates money to commodities through the labour
theory of value, while ignoring credit money. Ingham suggested instead that money
is constituted by social relations characteristic of credit, namely relations of ‘promise
to pay’. Drawing on Marx, this article shows that, pace Ingham, it is necessary
theoretically to relate money to commodities. Money is indeed constituted by social
relations, but these are commercial relations among ‘foreign’ commodity owners, not
credit relations. The social relations of money are successfully captured by Marx’s
concept of the universal equivalent, when that is interpreted as monopoly over the
ability to buy. In this light, both commodity and credit money are forms of the
universal equivalent, but qualitatively different from each other.

Keywords: Theory of money; Marx; barter; commodity exchange; labour theory of

value; post-Keynesianism.

Introduction: the social relations of ‘money in general’

The debate between Zelizer (2000) and Fine and Lapavitsas (2000) in the
pages of Economy and Society refers to the conceptualization of money.
Zelizer rejects the theorizing of money by neoclassical economics (and some
sociology), and claims that the concept of ‘money in general’ is invalid. In

Costas Lapavitsas, Department of Economics, School of Oriental and African Studies,

University of London, London, UK. E-mail: cl5@soas.ac.uk

Copyright # 2005 Taylor & Francis Group Ltd

ISSN 0308-5147 print/1469-5766 online
DOI: 10.1080/03085140500112137
390 Economy and Society

contrast, Fine and Lapavitsas defend the concept of ‘money in general’ and
analyse it from a Marxist perspective. Intervening in the debate, Ingham
(2001: 305) finds both sides in need of ‘untangling’, despite also ‘strongly
agreeing’ with Fine and Lapavitsas on the main issue in contention. In
particular, he criticizes Fine and Lapavitsas for drawing on Marx’s work,
which he considers incapable of supporting a theory of ‘money in general’.
Complicating things further, Ingham (2001: 305) also declares himself ‘at
odds with Fine and Lapavitsas’s interpretation of Marx’s conception of
money’. For Ingham, in short, Fine and Lapavitsas are right to stress the
importance of ‘money in general’ but wrong to rely on Marx, whom they
misinterpret to boot.
It is notable that Ingham rejects Fine and Lapavitsas’s analysis without
properly getting to grips with it. His true aim is to present an alternative
theory of ‘money in general’, associated with the German Historical
School and post-Keynesianism. Neglect of the alternative, apparently, led
Fine and Lapavitsas (and Zelizer) to focus excessively on commodity
money, while ignoring credit money and disregarding the social relations
inherent to money as ‘promise to pay’. The charge of neglect is surprising,
since one of the disputants has extensively discussed this alternative theory,
in places even mentioned in the original exchanges (Lapavitsas and
Saad-Filho 2000; Itoh and Lapavitsas 1999: chs 2, 10). To avoid covering
the same ground, therefore, Ingham’s alternative approach is recapitulated
in the ‘Money as a unit of account . . . ’ section of this article and
weaknesses are identified of which he seems unaware. These weaknesses
arise precisely because of denying the connection between money and
commodities.
The focus of this article, however, is on the positive aspect of Ingham’s
intervention, namely his claim that ‘money in general’ is ‘constituted by
social relations ’ (2001: 305, emphasis in original). This is an important
insight, except that the social relations involved are not those assumed by
Ingham (2001: 312, see also Ingham 1996, 1998, and, more fully, 2004a),
i.e. relations of ‘promise to pay’ or ‘credit
/debit’. Ingham is mistaken to
treat credit money as the generic type of ‘money in general’. Rather, the
social relations that constitute money are those among commodity owners
engaging in exchange. These relations unfold out of initial contacts
between commodity owners that take the form of ‘making a request for
exchange
/ receiving the ability to exchange directly’. Money subse-
quently monopolizes the ability to exchange directly (buy), and thus acts
as social nexus among commodity owners. The common content of
commodity and credit money is their absolute ability to buy
/ not some
imaginary ‘promise to pay’. This treatment of the social relations of
money relies heavily on Marx’s analysis of the universal equivalent. Far
from being irrelevant, Marxist political economy is a sine qua non for the
theory of ‘money in general’.
 Costas Lapavitsas: The social relations of money 391

Money and the labour theory of value

According to Ingham (2001: 314; see also Ingham 1998, 1999, 2004a: 61
/3)
the labour theory of value is the weak point of Marx’s theory of money:
‘Marx’s theory of money is flawed
/ like those of other classical economists

/ because it is grounded in the labour theory of value. . . . There is no
determinate link between money and commodities.’ Nonetheless, Ingham also
thinks that Fine and Lapavitsas have abused the ‘classical’ labour theory of
value:
However, in Fine and Lapavitsas’s interpretation of Marx, we are not offered the
classical labour theory of value nor a reference to any of the recent efforts to
reconcile classic Marxism with the reality of credit-money. . . . Rather, they
present an essentially Hegelian formulation of the origins of money.
(Ingham 2001: 315)
Ingham’s ‘classical’ (presumably ‘classical Marxist’) labour theory of value is
not at all clear, nor is it apparent how Fine and Lapavitsas diverge from it. Be
that as it may, it is shown below that the labour theory of value remains a vital
source of insights for the theory of money. The discussion draws on the work
of Fine and Harris (1979) and Weeks (1981) as well as Japanese Marxism of the
Uno tradition, especially Itoh (1976).
The distinction between form and substance of value is fundamental to
establishing the economic content of ‘money in general’. It is undeniable that
commodities possess the form of value, that is, they always exhibit quantitative
equivalences with each other
/ they have exchange value. For Marx, however,
commodities produced under capitalist conditions also contain definite
amounts of a social substance of value (abstract human labour). For
commodities produced by capital, value’s form is anchored on value’s
substance through a set of social and economic processes. These include the
elimination of non-transitivity among commodity prices through regular
buying and selling, the movement of workers indifferently between jobs and
the homogenization of work effort as workers are subjected to capitalist
exploitation at the workplace. They also include equalization of profit rates
across industrial sectors inducing market prices to gravitate towards prices of
production.
However, the form of value can also become detached, or even completely
divorced, from the substance of value. In capitalist economies, there is an
enormous array of things and activities that appear as commodities without
bearing any relation to produced commodities, for instance, real estate, shares,
insurance instruments, bribes, fines, and favours. Such things and activities
acquire the form of value (money prices) despite being only tenuously related
to value as abstract labour. By the same token, their prices heavily reflect non-
economic and arbitrary influences (psychological, political and institutional).
Moreover, in non-capitalist societies the form of value is also largely
unconnected to the deeper reality of production, since the substance of value
392 Economy and Society

is largely absent. Non-capitalist money prices acquire regularity and

transitivity due purely to market relations of demand and supply, backed by
the habits associated with repeated transactions.
Separation of form from substance of value implies that the economic
process of emergence of money is not connected to the substance of value.
Equivalently, money’s emergence is associated with the development of the
form of value (Lapavitsas 2003: ch. 3). This approach surprised Ingham (2001:
315), to whom it (erroneously) seemed ‘Hegelian’. On the contrary, it is both
materialist and Marxist because it shows money to be the outcome of social
relations among commodity owners. Fundamental to it is the assumption that
commodity owners approach each other as ‘foreign’ individuals. The term
‘foreign’ is used to denote the absence of pre-existing ties of kinship, hierarchy,
tradition and morality among commodity owners that might determine the
fundamental content of their exchanges. Commodity owners are disinterested
individuals who simply aim at obtaining an equivalent for what they bring to
market. Similarly, they do not even need to know each other or, in neoclassical
terms, the market is ‘anonymous’.
It follows that, at any random meeting of two ‘foreign’ commodity owners,
there must be an opening gambit that invites trading relations to occur. An
important theoretical innovation subsequently made by this approach to
money is to define the opening gambit in terms of the binary opposition of
‘relative-equivalent’ analysed by Marx (1976 [1867]: 139) in connection with
the ‘accidental’ form of value. Specifically, the opening gambit is taken to be a
request for exchange made by the relative party, offering own commodity for
the commodity held by the equivalent party.1 On this basis, an analytical
process is specified that leads to money’s emergence by drawing on the social
relations of ‘relative-equivalent’. Ingham appears to have been taken aback by
the originality of this treatment of money, the key components of which are
recapitulated below.

Social relations of the universal equivalent as monopolist of the ability

to buy

The opening gambit of making a request for exchange gives definite direction
to the ‘accidental’ relationship between two commodity owners. The relative is
the active party, whose request puts the other in the position of the equivalent,
or passive, party. In economic terms, the relative declares the exchange value of
own commodity to be represented by a quantity of the commodity possessed
by the equivalent. Simultaneously, the equivalent finds that own commodity
can be exchanged directly with (buy) that of the relative. This property of the
equivalent commodity is rudimentary ‘moneyness’, deriving purely from the
request for exchange made by the relative. Money as the universal equivalent
eventually monopolizes the ability to buy, due to spontaneous requests for
exchange made by all other commodity owners.
 Costas Lapavitsas: The social relations of money 393

Monopolization of the ability to buy occurs in successive (analytical) stages,

namely as ‘accidental’ exchange becomes ‘expanded’, then ‘general’ and, fin-
ally, ‘monetary’. The ‘expanded’ stage follows naturally from the ‘accidental’,
since each commodity owner could in principle address requests for exchange
towards any and all others. The ‘expanded’ stage captures the social relations
of one relative confronting endless equivalents, when commodity owners
regularly and frequently enter the process of exchange. The owners of the
equivalent commodities find that they have acquired a degree of buying ability,
if only towards a single relative. The ‘general’ stage, in contrast to the
‘expanded’, captures the reverse social relations, that is, of endless relatives
addressing a single equivalent. At the ‘general’ stage, all commodity owners
but one make regular and frequent requests for exchange to a single
commodity. The single commodity thus possesses overwhelming ‘moneyness’.
However, demonstrating the analytical passage from the ‘expanded’ to the
‘general’ stage is far from easy.
A key observation in this respect is that money represents extreme
asymmetry among commodities: one commodity is permanently placed on
the side of the equivalent and all others on the side of the relative. But
commodities are intrinsically symmetric as objects of trade, and this militates
against establishment of the absolute asymmetry that is characteristic of
money. A solution for this problem, developed elsewhere (Lapavitsas 2003):
ch. 3, 2005), is to take extra-economic forces, including social custom, as
fundamental to inducing monopolization of the ability to buy by money. But a
significant difficulty in this connection is that existence of social custom cannot
be immediately assumed among commodity owners, since they are essentially
‘foreign’ and hence unconstrained by kinship, religion, hierarchy and so on. If
commodity owners acquired customary relations, these would still be relations
among ‘foreigners’. A further difficulty is that the very existence of ‘foreign-
ness’, as defined above, seems particularly unlikely within non-capitalist
societies, in which economic activity is ‘embedded’ in power, prestige, kinship
and custom. On the other hand, ‘foreign-ness’ naturally prevails among
capitalist traders, and could also obtain at the points where non-capitalist
communities and societies engage in trade with each other. This is why Fine
and Lapavitsas refer to Marx’s claim (e.g. 1976 [1867]: 182, 1973 [1939]: 223,
1981 [1894]: 447
/8) that commodity exchange historically arises where
separate communities come into contact with each other. Ingham (2001:
316) dismisses this view as based on ‘long since superseded history’, but misses
its analytical importance and, as is shown below, grossly overestimates the
historical validity of his preferred alternative view.2
Within chains of customary transactions involving ‘foreigners’ who trade
specific commodities, conditions are likely to exist that lead to money’s
emergence. A commodity that is customarily and frequently traded could by
chance attract several requests for exchange, leading to the transient
appearance of the ‘general’ stage. This could occur for more than one
commodity within a chain of transactions. But if a commodity became a
394 Economy and Society

universal equivalent even temporarily, its enhanced ability to buy would

constitute an additional (exchange-related) use value, which Marx (1976
[1867]: 184) calls a ‘formal’ use value. Therefore, that specific commodity
would be likely to attract further requests for exchange, strengthening its
ability to buy and leading to still more requests for exchange. A process of
monopolization of buying power would be set in train. For the ‘money’ stage to
emerge properly, however, extra-economic factors are again necessary. The
physical properties of commodities are important in this respect, since
durability, homogeneity, divisibility and portability are desirable for the
monopolist of buying ability. Social custom is also important, since
commodities used for wealth display would be more naturally associated
with the ability to buy. Finally, after money has emerged, its continuous use
still relies on social custom. Commodity owners automatically offer their
commodities for money, expecting that they could obtain money in exchange.
The expectations of commodity owners are continually validated by their
collective practice.
To recap, Marx’s analysis of the form of value provides foundations for the
analytical derivation of ‘money in general’ as monopolist of the ability to buy.
The absolute asymmetry between money and commodities results partly from
economic processes and partly from non-economic relations, including social
custom. Money, thus, encapsulates the social relations of ‘foreign’ commodity
owners. This result stands in sharp contrast to Ingham’s view of money as
constituted by social relations characteristic of credit money and ‘promises to
pay’. The weaknesses of Ingham’s view are briefly summarized below.

Money as unit of account and credit money

According to Ingham (2001: 315), Marx has little to say on credit money and
‘seems to understand’ its peculiar character only in relation to pre-capitalist
formations. Marx’s view of capitalist credit instruments was ‘quite conven-
tional’ for his era, treating these as ‘substitutes ’ for hard cash (Ingham 2001:
315, emphasis in original). Ingham even places Marx in the same camp as
orthodox monetary theory as far as credit money is concerned, in opposition to
post-Keynesianism. But his claims are a caricature of Marxist theory of credit
money.
One of the historical antecedents of post-Keynesianism was the Banking
School, famous for its critique of the Bank Act of 1844 in England. The
Banking School had an elaborate theory of credit money, rejecting the notion
that banknotes (and deposits) were a mere substitute for hard cash (gold). They
also rejected the orthodoxy of the quantity theory of money, while putting
forth the ‘law of the reflux’ as regulating principle of the quantity of credit
money. There is no doubt that Marx was strongly sympathetic to the Banking
School.3 It is surprising to read Ingham’s assertion that ‘Marx held the
conventional contemporary Currency School view that credit instruments
 Costas Lapavitsas: The social relations of money 395

(bills of exchange, promissory notes, etc.) were, or rather should be, in a

rationally organised system, no more than functional substitutes for hard cash’
(2004a: 62). Entirely the reverse holds true. Marx differentiated between
commodity and credit money (Marx 1970 [1859]: 116); distinguished between
plain fiat ‘paper’ money and credit money (1976 [1867]: 224); postulated a
‘cyclical’ path for credit money, transparently similar to the ‘law of the reflux’
(1970 [1859]: 102, 1976 [1867]: 210); and showed disdain for the quantity
theory of money (1970 [1859], 1976 [1867]).
It is equally surprising for Ingham (2001: 316) to claim that Marxist and
orthodox monetary theory are similar because both, apparently, ‘ignore the
distinctiveness of capitalist banking’s creation of money through the act of
bank lending’. The affinity between Marxist and post-Keynesian analysis of
credit money creation through bank lending is acknowledged even by leading
post-Keynesian monetary theorists (theoretical differences notwithstanding)
(Lavoie and Secareccia 2001).
Ingham is nevertheless right to state that Marxist and post-Keynesian
monetary theories differ profoundly on the issue of money’s emergence and its
connection with commodity exchange. In a nutshell, the post-Keynesian view,
partly developed by Ingham, claims that money emerges as abstract unit of
account, typically in the realm of credit and through the action of an extra-
market authority, possibly the state. Ingham traces the theoretical roots of this
approach to the German Historical School and the Methodenstreit , but he is
unaware of its deficiencies. Distinctive views on money’s origin certainly
emerged with the German Historical School. Knapp’s (1924 [1905])
‘chartalism’ or ‘nominalism’, for instance, postulated that money is an
arbitrary quantification of purchasing power, a quantitatively specific material
claim on wealth. This was in opposition to Menger (1981 [1871], 1892), the
chief neoclassical opponent of the Historical School, who attempted to show
that money emerges spontaneously as means of exchange.4 Knapp’s arguments
influenced Keynes (1973 [1930]: 3) who claimed that money should be
theoretically understood as abstract unit of account for prices, debts and
contractual obligations, noted approvingly by Ingham (2001: 306, 1996, 1998,
2004a: 50
/2). Keynes further claimed that money as abstract unit of account is
of hoary antiquity, as proven by the ‘baked bricks’ of ancient Mesopotamia.
The influence of this argument on post-Keynesians has been strong, especially
on Wray (1990, 1998, 2000), on whom Ingham relies greatly (2001: 308
/10,
2004a: 52
/6). For Wray (1990: 13), ‘money in general’ appears to be credit-
money, i.e. debit and credit entries that allow transactions to proceed. This is
supposed to stand in contrast to neoclassical treatments of money, which focus
on the function of means of exchange and ignore the broader functioning of
money.
Ingham seems unaware of better-developed antecedents of his preferred
alternative approach to money. In particular, he offers no discussion of Sir
James Steuart, who systematically differentiated between ‘money of account’
(an arbitrary scale of value measurement) and ‘material money’ (money in
396 Economy and Society

actual use) (Steuart 1995 [1767], 2, iii: chs 1, 2). ‘Money of account’ functions
as abstract numeraire, while ‘material money’ generates practical approxima-
tions of abstract prices.5 For Steuart, ‘material money’ need not have the same
nomenclature as the abstract numeraire, since it is only an approximation of
the ‘money of account’. Moreover, actual prices established by ‘material
money’ need not coincide with ideal prices established by the ‘money of
account’. Thus, for Steuart, money is a social convention both as abstract unit
of value measurement and as means of exchange that approximates ideal prices
in practice.6 In postulating this distinction, Steuart gave theoretical form to the
mythical mercantilist stories of the macoute, i.e. the imaginary gold bar
presumably used by the natives of West Africa to measure commodity value.
The attempt to associate money’s origin with the social invention of an
abstract unit of account is beset with problems. Two of those are briefly
discussed and a third more fully explored below. First, and applying
particularly to post-Keynesians and Ingham, the postulated link between, on
the one hand, money operating as abstract unit of account and, on the other,
credit relations occurring among exchange participants, is extremely tenuous.
It is theoretically possible for money to operate as abstract numeraire in
ordinary commercial transactions that involve immediate exchange of
equivalents and have no connection with credit, as suggested by Steuart.
More significantly, operating as unit of account is neither the only nor even the
most important function of money in credit transactions. Equally fundamental
to credit is the operation of money as means of payment
/ unless the
outlandish assumption is made that all credit obligations are mutually cleared
or fresh credits are automatically extended at all times due. But when money
functions as means of payment in the settlement of debts, it automatically
functions as (broad) means of exchange, though not as (narrow) means of
circulation. Furthermore, to function as means of payment, money must be
able to preserve purchasing power, i.e. it must already function as hoard
element. In short, credit transactions are complex economic phenomena that
rely on the full panoply of money’s economic functions. Assigning exceptional
theoretical importance to money as unit of account in credit relations is
arbitrary and misleading.
Second, there is no unambiguous historical evidence of the existence of a
purely abstract unit of account, despite Ingham’s assertions (2001: 310; see also
2000). Merely showing that the unit of account happens to be differently
denominated from the means of exchange is no such evidence. The
performance of the two functions by differently denominated money during
any given period is one of the commonest features of monetary exchange. What
must be demonstrated is the existence of money of account that did not
originally function as means of exchange, i.e. money of account with purely
ideal units, products of human consciousness alone. In this respect, it is far
from sufficient to point at the existence of societies, such as in ancient Egypt
and Babylonia, which did not possess broad commodity exchange but in which
money functioned as unit of account in the royal or priestly economy. For
 Costas Lapavitsas: The social relations of money 397

these societies certainly traded, at the very least with foreigners, and
their money of account typically included standard quantities of a few
products. This is apparent in the edited volume by Wray (2004), which boldly
asserts the existence of ideal money of account in ancient Babylonia but only
succeeds in showing that these units of account were quantities of silver and
barley.
Finally, third, the hapless search for an ideal unit of account in history is
premised on theoretical confusion, apparent in Steuart’s argument summar-
ized above. There is no doubt that the accounting system of prices is an
abstract entity, which can be fully established on a sheet of paper by using
an ideal unit of money. It is a simple exercise in economics to generate such an
abstract nomenclature of price by using any number of different ‘numeraires’.
Moreover, in the practice of monetary exchange, the price of a specific
commodity is ideally determined in the mind of its owner prior to actual
exchange. In other words, money’s function as measure of value could
certainly be undertaken by purely imaginary or ideal money. Marx offers
strong insight on this issue:
Since the expression of the value of commodities in gold is a purely ideal act, we
may use purely imaginary or ideal gold to perform this operation. Every owner
of commodities knows . . . that it does not require the tiniest particle of real gold
to give a valuation in gold of millions of pounds’ worth of commodities. In its
function as measure of value, money therefore serves only in an imaginary or
ideal capacity. This circumstance has given rise to the wildest theories.
(Marx 1976 [1867]: 189
/90)

This aspect of the measure of value impressed Innes (1913, 1914), who is
acclaimed by Ingham and the post-Keynesians as a founding father of their
alternative approach to money. Innes subsequently turned the ideal measure-
ment of value into a ‘wild’ theory of all money as credit money:
The eye has never seen, nor the hand touched a dollar. All that we can touch or
see is a promise to pay or satisfy a debt due for an amount called a dollar . . . .
The theory of the abstract standard is not so extraordinary as it first appears,
and it presents no difficulty to those scientific men with whom I have discussed
the theory. All our measures are the same. No one has ever seen an ounce or a
foot or an hour.
(Innes 1914: 56)

However, in monetary exchange, ideal must also become actual price, if

commodity owners are actually to obtain the equivalent they ideally expect
(and if debt holders are to receive value due that has not been cleared against
other debt). Money acts initially as ideal measure of value, but if exchange is to
have economic content, money must also act as standard of price in practice,
thereby rendering prices real. The ideal measurement of value is only a first
step in the process of exchange (including commodities and debts)
/ at some
398 Economy and Society

point value must also be measured in practice and rendered into actual price.
Evidently, the transformation of ideal into actual prices has nothing to do with
ideal money units, and depends entirely on actual money. In this respect,
the money stuff and the denominations of actual money are of the first
importance. Thus, a given set of commodity values might be ideally measured
in gold, but would produce very different actual prices for gold denominated in
sovereigns, francs, dollars and so on. The actual price system would be
different still if value continued to be measured ideally in gold but actual
money comprised silver or bronze, variously denominated.
Therefore, money functioning as measure of value should be differentiated
from money acting as standard of price, as Marx noted (also associating
measure of value with abstract labour):
As the measure of value it [money] serves to convert the values of all the
manifold commodities into prices, into imaginary quantities of gold; as the
standard of price it measures those quantities of gold. The measure of value
measures commodities considered as values; the standard of price measures, on
the contrary, quantities of gold by a unit quantity of gold, not the value of one
quantity of gold by the weight of another.
(Marx 1976 [1867]: 192)

An important point here is that the denominations of the standard of price

are social conventions. This is not simply because the state, or some other
authority, dictates the standard of price, as has commonly been the case
throughout history. Rather, the conventional aspect of the standard of price is
associated with the physical aspect of the money stuff as well as the social
customs that surround its use as plain commodity: salt comes in blocks that
vary with production method and social custom of use, cloth is cut in standard
measures that differ according to material and habits of use, cattle is counted in
heads, precious metals of variable fineness and standard weight are used by
different communities according to their own traditions. The ability of the
state to impose on society its own units of money of account rests on the
socially conventional nature of the standard of price. The state gives order to
the conventions surrounding the standard of price partly for reasons of its own
(tax and seigniorage) and partly to lessen the inevitable frictions arising from
several standards operating concurrently. This is apparent in the case of
coinage, but not substantially different when the state chooses as standard of
price a particular banknote, or an arbitrary unit of fiat money which is loosely
and indirectly connected with the money commodity.
In this light, the alternative approach advocated by Ingham confuses the
undoubted ability of the state (or another socially constituted authority) to set
the standard of price with an (imaginary) ability arbitrarily to set the measure
of value. The state can create its own price numeraire, but this is because a
spontaneous measure of value already exists that is conventionally denomi-
nated as standard of price and acts as means of exchange. The abstraction
 Costas Lapavitsas: The social relations of money 399

involved in creating a price numeraire (by the state or another authority) refers
to determining the standard of price and not the measure of value. Money as
the measure of value is not abstract (and far less ideal)
/ it is one commodity
emerging spontaneously among the many. Adam Smith, for all the criticism to
which he has been subjected in this respect, including by Ingham (2004a: 34),
was right in refusing to adopt Steuart’s theory of the abstract measure of value,
despite being perfectly familiar with it.
Precisely because money is not an ideal unit of value measurement (except
in general equilibrium exercises by economic theorists), Ingham’s favoured
approach faces insuperable difficulties in developing a logical account of how
such an ideal unit could have been devised in practice. In this respect, this
theoretical current offers nothing remotely comparable to Marx’s dialectical
analysis, or Menger’s taut derivation of money’s emergence. Innes (1913,
1914), for instance, gives no logical account of how the putative abstract unit of
value measurement emerged. We are left with the vague supposition that the
machinery of ancient states somehow devised a coherent system of values by
fixing an abstract unit of account. Political economy, meanwhile, was able to
perform this gigantic feat of abstraction only after centuries of effort and
against the concrete reality of constant commensuration of commodities in
capitalist markets.
Ingham (2004b: 175
/83), to his credit, realizes Innes’s weakness in
this respect. Thus, he searches for an answer and finds it in the numismatic
work of Grierson (1977) (Ingham 2001: 310, also 1996: 519
/21, 2004a: 90
/3,
2004b: 182
/3). Grierson suggests, along lines earlier explored by the German
Historical School, that money emerged as unit of account in wergeld , that
is, in the practice of making monetary compensation for social and individual
‘injuries’. Grierson is led to this argument because he thinks that it
is impossible to render commensurate any significant number of commodities,
in view of the huge number of bilateral combinations thereby generated.
Thus, he assumes that a more plausible path toward commensuration of
disparate things was provided by communal assemblies that determined
equivalences among a ‘few’ injuries for the purpose of compensation. In this
light, the abstract unit of money was communally devised in the practice
of wergeld .
Grierson’s erudite work is not persuasive in this regard. For one thing,
commodity exchange and the use of money probably preceded the practice of
wergeld in history. More significantly, Grierson shows little appreciation of the
process of value measurement in the realm of commodities, and for this reason
thinks of the large number and the inherent natural differences of commodities
as forbidding obstacles to commensuration. But these putative weaknesses are
actual strengths of the spontaneous process of commodity commensuration in
the course of exchange. Large numbers of repeated bilateral transactions
(premised on differences among use values) are necessary to establish relative
prices as objective social norms, while providing the necessary impetus for
monopolization of buying ability by money. It is more likely, indeed, that the
400 Economy and Society

problem holds in reverse. Grierson’s assumption that injuries could be brought

into equivalence more easily than commodities is far from obvious, in view of
the moral, familial, customary, hierarchical, ethical and other factors at play.
Moreover, even a short list of injuries would generate a large number of
bilateral relations that would have to be reassessed whenever new ‘injuries’
emerged or old ones dropped out. The task for communal assemblies in
devising a consistent abstract standard of measurement, and thus compensa-
tion, would be Herculean.
In sum, therefore, a theory of ‘money in general’ that is unrelated to
commodity exchange is unsatisfactory. Its weaknesses do not necessarily imply
the need to adopt the neoclassical approach, which simply inserts a convenient
means of exchange into barter. The Marxist analysis that puzzled Ingham
avoids these difficulties by showing that money emerges as monopolist of
buying ability.

Conclusion: credit money, commodity money and ‘money in general’

The complex issues raised in this article do not lend themselves to a short
conclusion. But it is instructive to note some analytical implications for the
relationship between commodity and credit money, believed by Ingham to
constitute an intractable problem for Marxist monetary theory.
Ingham (2001: 316, also 1998, 2000) repeatedly asserts that Marx
believes money to be ‘essentially’ a commodity, and hence seeks money’s
origin in trade. It is clear, in view of the discussion above, that this is an
erroneous assertion. For Marx, money is the universal equivalent or
independent form of value. There is no reason to assume that the universal
equivalent is ‘essentially’ a commodity. On the contrary, it can take a variety of
forms
/ commodity, fiat paper, banknotes, bank deposits, money trust
accounts, and so on. The commodity form of money is certainly fundamental,
not least because it is the form in which money originally emerges in
commodity exchange. But none of money’s forms has exclusive rights to
representing money’s ‘essence’. Rather, in all its forms, the universal
equivalent remains the monopolist of the ability to buy, this being the thread
that binds its forms together.
Indeed, a far more serious problem in this respect can be found in Ingham’s
work (1996, 2004b). In his effort to show that money is essentially a ‘social
relation’, Ingham finds it necessary to denounce any suggestion that money is a
‘thing’. This is misleading. Money is a social relation among ‘foreign’
commodity owners, while also being a thing. Social relations embodied in
things are characteristic of capitalist markets, analysed by Marxist theory in
terms of commodity fetishism. The thing-like aspect of money is particularly
obvious in situations of economic crisis, during which a capitalist’s social
relations might remain completely unchanged, but bankruptcy could easily
 Costas Lapavitsas: The social relations of money 401

occur due to temporary lack of the money ‘thing’. Much of the mystery and
complexity of money arise because it is simultaneously a social relation and a
thing. The particular form taken by the money ‘thing’, moreover, is important
for money’s functioning.
There are significant economic differences among the forms of money.
Capitalist commodity money, for instance, inherently contains value (abstract
labour). This alone shows the fallacy of Ingham’s main claim, namely that
there exist ‘generic social relations of the system of promises to pay’ which
apply to all forms of money (2001: 307, emphasis in original). Capitalist
commodity money is no-one’s liability and bears no necessary relation to credit
processes: it is not a promise to pay but the instrument of final payment.
Credit money, on the other hand, is indeed a promise to pay, but this
constitutes its qualitative difference with commodity money. It is plain
confusion to lump together commodity and credit money as ‘promises to
pay’. There is a common aspect to all forms of money but this is their unique
ability to buy rather than some fictitious ‘promise to pay’. To establish this
result, however, it is necessary to relate money to commodities, precisely the
approach that Ingham advises against.
Credit money presents no exceptional difficulties for Marxist monetary
theory.7 Money as monopolist of buying ability necessarily functions as means
of payment by separating purchase from sale, thus giving rise to trade credit
relations. Money also makes possible the transfer of buying power among
capitalists through relations of lending. Credit relations of both types are
continually generated by private capitals and proliferate in a capitalist
economy. Credit money emerges in credit transactions among commercial
and industrial capitals, acquires more developed forms in the operations of
banks, and becomes the dominant form of money. Capitalist credit money
certainly comprises promises to pay but that is also why it differs from
commodity money: financial institutions systematically create and eliminate
credit money by issuing and settling promises to pay. Nevertheless, credit
money remains a form of the universal equivalent, sharing in common the
character of ‘money in general’ that is associated with the social relations of
commodity exchange.

Notes

1 The opening gambit is discussed in detail in Lapavitsas (2003: ch. 3, 2005), where it
is also treated as rudimentary offer to sell.
2 This is more fully discussed in Lapavitsas (2003: ch. 3).
3 See Lapavitsas (1994, 1996).
4 The two approaches have been known as, respectively, the acatallactic and the
catallactic theory of money’s origin (Mises 1953 [1934]). In Lapavitsas (2003: ch. 6)
they are called ‘money as unit of account’ and ‘money as means of exchange’, and both
are discussed from a Marxist standpoint.
402 Economy and Society
5 Ingham (2001: 310) draws a distinction between a ‘monetary’ (‘abstract’) duck and a
‘commodity’ (concrete) duck, in any ‘duck standard’ of value. This is strongly
reminiscent of Steuart.
6 Marx (1970 [1859]: 79
/81), who had a high regard for Steuart’s monetary theory,
differed from him on this issue (Itoh and Lapavitsas 1999: ch. 1).
7 See Lapavitsas (1991, 2000).
References
Fine, B. and Harris, L. (1979) Rereading
Capital , New York: Columbia University
Press.
Fine, B. and Lapavitsas, C. (2000)
‘Markets and money in social science:
what role for economics?’,
Economy and Society 29(3): 357
/82.
Grierson, P. (1977) The Origin of Money,
London: Athlone Press/University of
London.
Ingham, G. (1996) ‘Money is a social
relation’, Review of Social Economy 54(4):
507
/29.
** / / (1998) ‘On the underdevelopment
of the ‘‘sociology of money’’’, Acta
Sociologica 41(1): 3
/18.
** / / (1999) ‘Capital, money and
banking: a critique of recent historical
sociology’, British Journal of Sociology
50(1): 76
/96.
** / / (2000) ‘‘‘Babylonian madness’’: on
the historical and sociological origins of
money’, in J. Smithin (ed.) What is
Money? , London and New York:
Routledge.
** / / (2001) ‘Fundamentals of a theory of
money: untangling Fine, Lapavitsas and
Zelizer’, Economy and Society 30(3):
304
/23.
** / / (2004a) The Nature of Money,
Cambridge: Polity Press.
**
/ / (2004b) ‘The emergence of
capitalist credit money’, in Wray (2004).
Innes, A. Mitchell (1913) ‘What is
money?’, Banking Law Journal , May:
377
/408, reprinted in Wray (2004).
** / / (1914) ‘The credit theory of
money’, Banking Law Journal , January:
151
/68, reprinted in Wray (2004).
Itoh, M. (1976) ‘A study of Marx’s theory
of value’, Science and Society 40(3):
307
/43.
** / / and Lapavitsas, C. (1999) Political
Economy of Money and Finance, London:
Macmillan.
Keynes, J.M. (1973 [1930]) A Treatise on
Money, Royal Economic Society, London:
Macmillan.
Knapp, G. (1924 [1905]) The State
Theory of Money, London: Macmillan.
Lapavitsas, C. (1991) ‘The theory of
credit money: a structural analysis’,
Science and Society 55(3): 291
/322.
** / / (1994) ‘The banking school and
the monetary thought of Karl Marx’,
Cambridge Journal of Economics 18(5):
447
/61.
** / / (1996) ‘The classical adjustment
mechanism of international balances:
Marx’s critique’, Contributions to Political
Economy 15: 63
/79.
** / / (2000) ‘Money and the analysis of
capitalism: the significance of commodity
money’, Review of Radical Political
Economics 32(4): 631
/56.
** / / (2003) Social Foundations of
Markets, Money and Credit , London:
Routledge.
** / / (2005) ‘The emergence of money in
commodity exchange, or money as
monopolist of the ability to buy’, Review
of Political Economy forthcoming.
** / and Saad-Filho, A. (2000)
/
‘The supply of credit money and capital
accumulation: a critical view of
post-Keynesianism’, Research in Political
Economy 18: 309
/33.
Lavoie, M. and Seccareccia, M. (2001)
‘Post-Keynesian and Marxist economics:
twins or distant cousins?’, paper
presented at the Progressive Economics
Forum, Canadian Economics Association,
May
/June, MacGill University,
Montreal.
 Costas Lapavitsas: The social relations of money 403

Marx, K. (1970 [1859]) Contribution to a and Chronological, of the Late Sir James
Critique of Political Economy, Moscow: Steuart , London: Routledge.
Progress. Weeks, J. (1981) Capital and Exploitation ,
** / / (1973 [1939]) Grundrisse, London:

Penguin/NLR.
London: Edward Arnold.
Wray, R. (1990) Money and Credit in
** / / (1976 [1867]) Capital , Vol. 1,

London: Penguin/NLR.
Capitalist Economies, Aldershot and
Brookfield: Edward Elgar.
** / / (1981 [1894]) Capital , Vol. 3,

London: Penguin/NLR.
** / / (1998) Understanding Modern

Money: The Key to Full Employment

Menger, K. (1892) ‘On the origin of and Price Stability, Cheltenham: Edward
money’, Economic Journal 2: 239
/55. Elgar.
** / / (1981 [1871]) Principles of

Economics, New York and London: New

** / / (2000) ‘Modern money’, in

J. Smithin (ed.) What is Money? , London

York University Press. and New York: Routledge.
Mises, von L. (1953 [1934]) The Theory
of Money and Credit, Appendix A: ‘On the
** / / (ed.) (2004) State and Credit

Theories of Money: The Contributions of

Classification of Monetary Theories’ ,
London: Cape.
Steuart, J. (1995 [1767]) An Inquiry into
the Principles of Political Economy, Vols 1,
2, 3, 4, in Works, Political, Metaphysical,
A. Mitchell Innes, Cheltenham: Edward
Elgar.
Zelizer, V. (2000) ‘Fine tuning the
Zelizer view’, Economy and Society 29(3):
383
/9.

Costas Lapavitsas is Reader in Economics at the School of Oriental and

African Studies, University of London. He works on political economy of
money and finance, and on the Japanese economy.